B.W.P.A. ANNUAL CONVENTION, 1984

Transcription

1 B.W.P.A. ANNUAL CONVENTION, 1984 WEATHERING INTERACTIONS ON TREATED AND UNTREATED WOOD SURFACES by W. C. FEIST U.S. Department of Agriculture, Forest Service, Forest Products Laboratory*, Madison, Wisconsin The primary objectives of the studies described in this report were to study the mechanism of outdoor weathering and document the weathering performance of representative exterior finishing systems on various wood and wood-based substrates representing a selection of currently available traditional and newer siding materials. The effects of pretreatment/ finish/substrate interactions and ultraviolet light-blocking characteristics were emphasised. These studies are currently in progress and much of the data represents early findings of long-range exposure tests. Solid wood and wood-based (composite) panel products represent well over half of the exterior siding (cladding) material used for dwellings and other structures in the United States. In 1982, 191,100,000 m 2 of wood, hardboard (fibre building board) and plywood siding were used on exterior walls (Table 1). These two panel siding products have dominated the siding market for some years in the United States, while the use of solid wood siding has been slowly declining. Many new panel products made from reconstituted wood materials are being introduced in the United States and the world market. Hardboard Plywood Redwood Cedar Others (Aluminium, Vinyl, Steel, Brick) Total TABLE 1 Exterior siding (cladding) in the United States 1 Product shipments surface basis Million m Services: American Hardboard Assoc., American Plywocd Assoc., Western Wood Products Assoc., U.S. Dept. of Commerce Additional wood products used as exterior-siding include solid lumber, waferboard, plywood composites, particleboard composites and others. All these wood and panel siding materials are usually finished in some way (paint, stain, natural finish) for outdoor weathering protection and for esthetic effects 3,9,17,19,27,33,37. A wide variety of finishes are applied to wood and woodbased siding materials 12,17,19,29,22,23,27,29,37. Semitransparent penetrating stain finishes are designed to emphasise wood grain and texture characteristics 3,8,23,26,29,31. Solid colour stains (also called opaque, hiding, or heavy-bodied stains) hide wood grain and colour while letting texture show 17,19,22,37. These finishes perform essentially like paints. Two- and threecoat house paints form a highly durable coating in addition to *The Forest Products Laboratory is maintained in cooperation with the University of Wisconsin. hiding grain and most surface texture. These finish systems offer various degrees of protection for wood and wood panel products against outdoor weatheting 9,12,20,27,32,33. Transparent film-forming finishes, such as varnish, provide only minimal protection to exposed wood surfaces 1,2,4,22,32. Several publications have described the use of water repellents as pretreatments for improving the weathering performance of wood and wood-based products 5,6,11,18,24,25,30,32,34. The value of the pretreatment of solid wood products with certain inorganic salts, notably those containing hexavalent chromium, has been clearly demonstrated 7,21,22,23,28. Many publicautions have described the weathering performance of a range of finishes on reconstituted wood panel products such as plywood, hardboard and particleboard 9,10,13,14,15,16,19,20,22,28, 30,34,35,36. Several review papers on wood finishing and weathering have described past work 9,19,22,33. Despite this relatively large research effort, a great deal of misunderstanding and many misconceptions exist about the proper finish and the best finishing practices for the individual wood-based products. There is relatively little comparative information about the performance of finish systems or about finish/substrate interactions. Additional complications are caused by the wide range of new finishing materials being applied to the wide range of new wood-based substrates. Problems with substrate/finish durability sometime result with these new materials. Finish durability problems may also be further compounded by environmental and governmental restrictions on traditional paint and stain systems, as well as by restrictions on some paint additives such as certain heavy metals, mildewcides, and preservatives. EARLIER WORK In a previous paper 20, we studied a number of commercial finishes on several representative wood siding products. The outdoor exposure studies in Madison, Wis., on the performance of several finish systems on four wood-based panel products and one solid wood substrate showed that two- and three-coat finish systems gave enhanced protection. The four panel products evaluated (hardboard siding, roughsawn redwood plywood, smooth Douglas-fir plywood, and aspen waferboard) exhibited different degrees of overall performance (Fig. 1). Aspen waferboard had the poorest overall performance, primarily because of particle delamination and the presence of white-rot decay organisms. The best overall performance was found for hardboard siding, especially for opaque paint finish systems. A semitransparent oil stain offered the least protection of all finish systems studied, because of its partial transparency to ultraviolet light. Good overall performance of all-acrylic latex finish systems on all substrates was clearly demonstrated. This earlier study was preliminary and prompted many of the studies currently under way. MATERIALS AND METHODS USED IN OUTDOOR EXPOSURE STUDIES At the Forest Products Laboratory, we have initiated a series of outdoor exposure studies. The following sections describe the general methodological similarities of our studies. Major exceptions to these methods are discussed in relation to each individual study.

2 14 RECORD OF THE 1984 ANNUAL CONVENTION OF THE BRITISH WOOD PRESERVING ASSOCIATION Wood substrates Substrates chosen for these outdoor exposure studies are shown in Table 2. The unprimed hardboard (HB), roughsawn plywoods and smooth or scratch sand plywoods represent materials currently used in large quantities for siding and other exterior wood surfaces such as soffits and fences. The aspen (Populus tremuloides) waferboard (WB) is finding increased use for structural sheathing and has been promoted for possible siding application. The sweetgum (Liquidambar styraciflua) roughsawn plywood (SGP-R) was included in our studies because newer alternate species (other than the traditional softwoods of cedar, redwood, etc.), particularly hardwoods, are being considered for siding. The western redcedar (Thuju plicata) (C) and southern pine** (Pinus sp.) (SP) bevel TABLE 2 Substrate characteristics for outdoor exposure performance studies Substrate Surface texture Thickness cm Western redcedar bevel siding (C) 1 3 Southern pine bevel siding (SP) Smooth planed 1 3 smooth 1 1 Simulated wood grain 1 0 Roughsawn 1 6 Smooth sanded 1 6 Roughsawn 1 6 Scratch sanded Aspen waferboard (WB) Hardboard (HB), unprimed Douglas-fir plywood (DFP-R) Douglas-fir plywood (DFP-S) Southern pine plywood (SPP-R) Southern pine plywoud (SPP-S) Western redcedar plywood (CP-R) Sweetgum plywood (SGP-R) 1 Roughsawn surface, lightly sanded afterwards Roughsawn 1 3 Roughsawn 1 6 **Southern pine includes the following species: Longleaf (P. palustris). Shortleaf (P. echinata). Loblolly (P. taeda), and Slash (P. elliottii). siding were included for comparative purposes. C represents a wood with excellent finishing and performance characteristics and SP, a wood that does not usually exhibit good performance 20,37. Table. 2 also shows abbreviations for each substrate that will be used in the remainder of this report. In the preparation of test panels, single 1 2 by 2 4 m sheets of each panel substrate were cut into eight 34 by 38 an specimens. Bevel siding specimens were cut into 38 an lengths from 2 4 m-long boards for incorporation into test panels having three boards per panel 20. Outdoor Exposure Panels All wood substrates were fastened to 41 cm wide by 34 cm long frames of 6 4 mm exterior grade plywood with 1 3 cm wide by 2 5 cm deep wood side rails. The frames were dip-treated with a water-repellent preservative 24, and edge-coated with an exterior acrylic latex house paint before test substrates were attached with stainless steel nails. The bevel siding was lapped 1 3 cm. Each frame held only one specimen of each panel substrate and three of each solid wood substrate. The test frames were hung on vertical fences with southern exposure at Madison, Wis. (latitude 43 N). Finishes and Pretreatments We selected a wide range of commercially available and laboratory prepared finishes and pretreatments for these studies (Table 3). Finish systems comprised of combinations of the individual finishes and/or pretreatments were used as well as individual materials. The materials selected represent finishes currently available (or recommended) for application on wood used in outdoor exposures. All finishes and pretreatments were brushed on the surface of the wood substrate under ideal laboratory conditions and following all the recommendations provided by the manufacturers, where applicable. Top, side, and bottom edges of all test panels were sealed with the finish or pretreatment itself. Spreading rates were those usually recommended 20 and were

3 WEATHERING INTERACTIONS ON TREATED AND UNTREATED WOOD SURFACES 15 TABLE 3 Finishes and pretreatments used in outdoor exposure studies Finish or pretreatment Solids Source Colour contents (%) Tung oil Linseed oil Oiticica oil Linseed oil + tung oil + phenolic resin Polyurethane varnish Water repellent Water-repellent preservative (solvent borne) Water-repellent preservative (waterborne) Pigmented shellac primer Latex pigmented shellac primer Alkyd primer paint Acrylic latex primer paint Acrylic latex house paint A Acrylic latex house paint B Alkyd house paint Semitransparent oil stain Semitransparent latex stain Solid colour oil stain Solid colour latex stain Chromium trioxide Organochrome complex 1 Turns green following application to wood Laboratory Laboratory Laboratory Laboratory Transparent 50 Transparent 50 Transparent 28 Transparent 25 Transparent 53 Transparent 13 Transparent 16 Transparent 4 White 54 White 52 White 78 White 52 White 52 White 53 White 62 Brown 75 Brown 25 Light brown 60 Light brown 45 Orange 1 5 Green 12 determined by direct weighing. All substrate surfaces were wiped with a soft cloth before finishing and between coats. No other special surface preparation was used. All substrates were stored indoors at 80 per cent RH and 26 7%. AU surfaces represented new, unexposed wood. All panels were allowed to dry for at least two days in the laboratory before being installed outdoors on the vertical test fence. Finish and Substrate Performance Ratings We used different criteria to determine the overall performance ratings 20 of the various pretreatment/finish/substrate systems (Table 4). Most of the evaluation methods were based on American Society for Testing and Materials (A.S.T.M.) Standards. Both substrate and finish performance were considered. The performance criteria used depended on the substrate type or finish type. TABLE 4 Inspection criteria and methods of evaluating performance Inspection criteria Evaluation method 1 Substrate integrity Checking ASTM 2 D Cracking, flaking. delamination, Discoloration deformation Subjective visual assessment Subjective visual assessment similar to ASTM D Mildew ASTM D Finish or surface evaluations Checking ASTM D Cracking ASTM D Erosion ASTM D Flaking ASTM D General appearance Subjective visual assessment 1 All evaluations used a 10 (perfect) to 1 (failure) scale 2 American Society for Testing and Materials Discoloration (unevenness of colour) from any cause and mildew were considered indicators of performance because of their visual effect. Substrate integrity was evaluated considering the individual characteristics of each substrate. The elements included (where applicable) erosion, flaking, peeling, cracking, and checking. Finish performance was evaluated for checking, cracking, erosion and flaking, where applicable. A general appearance rating (subjective visual assessment) was also used as a final overall criterion. Thus, the overall performance rating was based on an average of the various elements of discoloration, mildew, substrate performance, finish performance, and general appearance of the system, each evaluated on a 10 to 1 scale. A value of 10 indicated no change from the original unweathered condition, 5 represented an overall condition at which refinishing would be required but without extensive preparation of the substrate surface, and 1 represented total failure. Completely objective rating observations are difficult to make. For consistency, observations were made by the same person on each occasion, and photographs were used to compare results from year to year. Evaluations were made annually. RESULTS AND DISCUSSION The following results from our outdoor exposure studies illustrate the performance of a wide range of commercially available finishes on a number of wood siding products currently used in the United States. The effectiveness of commercial pretreatments (primarily water-repellent preservatives) and some laboratory pretreatments in improving finish and substrate performance are illustrated in each of the studies as appropriate. The finish systems chosen for these studies represent a range of systems of recognised low durability or long-term durability (i.e., systems that are known to last on solid wood only a year or two to those known to last up to 10 years 18,19 ). Penetrating Oil Finishes The results of a small study of penetrating oil finishes clearly illustrate that penetrating unpigmented oil finishes have only a short term durability on outdoor exposed wood surfaces (Table 5). Because exterior plywood should be protected from sunlight and substrate surface checking should be minimised, these oil finishes are not suitable for those products 13,15,16,28 formulations can be expected to give somewhat better protection than oil finishes, but these formulations would still not be suitable for plywood.

4 16 RECORD OF THE 1984 ANNUAL CONVENTION OF THE BRITISH WOOD PRESERVING ASSOCIATION TABLE 5 Performance of penetrating oil finishes on wood siding products after 1 year of outdoor exposure Western redcedar beveled siding Douglas-fir roughsawn plywood Sweetgum roughsawn plywood (C) (DFP-R) (SGP) Wood finish Surface Dis- Surface Dis- Surface Discheck- color- Mildew General check- color- Mildew General check- color- Midew General ing ation ing ation ing ation Tung oil Linseed oil Oiticica oil Linseed oil + Tung oil + Phenolic resin Water-repellent Preservative i Contained 0 5% 3-iodo-2-propynl butyl carbamate Finishes based on drying oils (linseed, tung, etc.) dissolved in suitable solvents have long been used as wood finishes because of their aesthetic appeal and ease of refinishing. The penetrating oil finishes are not pigmented and are considered the most natural of all wood finishes, highlighting the grain and figure of the wood surface. Unfortunately, these penetrating oil finishes do not provide any long-term protection to wood. In this study, exposed surfaces of western redcedar (C), Douglas-fir roughsawn plywood (DFP-R), and sweetgum roughsawn plywood (SGP-R) developed surface checking and roughening after only one year of outdoor exposure (Table 5). Discoloration and mildew growth was strongly evident on the exposed wood surfaces and was much worse than on the untreated control specimens. The overall general appearance of these penetrating oil-finished wood substrates had deteriorated quite rapidly during the one-year exposure. The oiticica oil and linseed oil/tung oil/phenolic resin formulations were generally better on C and DFP-R than on the SGP-R. The best overall natural finish was the laboratoryprepared water-repellent preservative () 24. This illustrates the principles of the need for a combination of sealing, water repellency and preservative action for a successful natural wood finish. A good water repellent () will minimise checking, cracking and deformation, and a good preservative (mildewcide) will control mildew (and other standing organisms), the primary cause of wood surface TABLE 6 Performance of paint systems on wood panel products pretreated with water-repellent preservatives after 4 years of outdoor exposure Paint system Substrate Finish Overall performance rating after checking/ cracking Flaking Cracking General 4 Years 3 Years 2 Years 1 Year Alkyd primer + alkyd topcoat + alkyd primer + alkyd topcoat (AQ) + alkyd primer + alkyd topcoat Latex primer + latex topcoat B + latex primer + latex topcoat B (AQ) + latex primer + latex topcoat B Alkyd primer + latex topcoat A + alkyd primer + latex topcoat A (AQ) + alkyd primer + latex topcoat A Alkyd primer + alkyd topcoat + alkyd primer + alkyd topcoat (AQ) + alkyd primer + alkyd topcoat Latex primer + latex topcoat B + latex primer + latex topcoat B (AQ) + latex primer + latex topcoat B Alkyd primer + latex topcoat A + alkyd primer + latex topcoat A (AQ) + alkyd primer + latex topcoat A Alkyd primer + alkyd topcoat + alkyd primer + alkyd topcoat (AQ) + alkyd primer + alkyd topcoat Latex primer + latex topcoat B + latex primer + latex topcoat B (AQ) + latex primer + latex topcoat B Alkyd primer + latex topcoat A + alkyd primer + latex topcoat A (AQ) + alkyd primer + latex topcoat A

5 WEATHERING INTERACTIONS ON TREATED AND UNTREATED WOOD SURFACES 17 TABLE 7 Role of pretreatment/primer combinations on the performance of paints after 3 years of exposure Finish system Finish performance Overall performance rating after Pretreatment Primer coat Paint topcoat Flaking Cracking General 3 years WESTERN REDCEDAR, SMOOTH BOARD (C) DOUGLAS-FIR ROUGHSAWN PLYWOOD (DFP-R) TEXTURED HARDBOARD (HB)

6 18 RECORD OF THE 1984 ANNUAL CONVENTION OF THE BRITISH WOOD PRESERVING ASSOCIATION discoloration in natural finishes 19,23,24,33. The excellent performance of the on the hardwood plywood, SGP, was most unexpected. Water Repellent Pretreatments on Wood Substrates When three different wood and wood panel siding products were pretreated with either a solvent borne or a waterborne one, and subsequently finished with one of three paint systems, some interesting, if not unexpected, results were observed (Table 6). The water repellent () or water-repellent preservative () treatment has had a long history as a pretreatment for wood exposed outdoors 5,11,18,19,24,34,36,37. These treatments have usually been applied to solid wood products before painting. There is little published information on the use of these pretreatments on composite wood panel products. In addition, the s are now available in both solvent borne (conventional) and waterborne (new) formulations. There is little published information on the performance of the newer waterborne formulations on wood siding. The three most common paint systems used on exterior wood siding are the following: alkyd primer paint plus alkyd topcoat paint; alkyd primer paint plus latex topcoat paint; latex primer paint plus latex topcoat paint. The all-acrylic all-latex paint systems are considered as the best overall paint systems for wood and wood panel products because of their ability to retain flexibility over long exposure times 16,20. They are especially useful on plywood panels because they help control or minimise face checking 16. In this study, the alkyd primer/alkyd topcoat system performed poorly on all substrates when the wood was not pretreated with (Table 6). Pretreatment with the solvent borne (mineral spirits solvent) greatly improved the performance of the ail-alkyd system. The waterborne was not as effective as a pretreatment for this paint system as was the solvent borne. The alkyd primer/latex topcoat paint system was superior to the all-alkyd system even on wood not pretreated with a. The pretreatment with solvent borne improved paint performance. The waterborne pretreatment was somewhat effective on the southern pine (SP) board, very effective on the southern pine scratch sanded plywood (SSP-S) and ineffective on the Douglas-fir smooth plywood (DFP-S). The all-latex system (based on all-acrylic paints) was generally the best overall paint system on all three substrates even when no pretreatment was used. An improvement was found in paint performance when the wood was pretreated with the solvent borne but not with the waterborne. The study summarized in Table 6 also included other substrates (C, HB, CP-R) and three-coat paint systems, but the data are not shown. The finishes and the substrates all performed well and the three-coat (one primer + two topcoats) paint systems, with or without pretreatment showed excellent durability after four years of outdoor exposure. In another related outdoor exposure study, the role of or pretreatments in combination with several primer/ paint combinations, was investigated (Table 7). Generally good to excellent performance was found for all the pretreatment/primer/paint systems studied with the striking exception of the system using or pretreatment followed by a pigmented shellac primer (alcohol solvent) and a quality acrylic latex house paint. There was some negative interaction with the system and the primer. Severe loss of adhesion of the paint resulting in flaking was found especially on the hardboard (HB) and on the C. After three years of outdoor exposure there were no striking differences between the performances of the other paint systems on the substrate evaluated whether untreated or pretreated with or. When the latex housepaint A was used as a self-primer and a topcoat on DFP-R. the or pretreatment had a positive effect. A similar positive effect was found when the pretreatments were used with oil primer and the same latex topcoat paint on DFP-R. Inorganic Salt Pretreatments This pretreatment study illustrates that an effective pretreatment like can greatly improve even the performance of semitransparent finishes on hardboard (HB). The improvement in performance probably results from the same mechanisms as those observed for solid wood products 21,22. Finish performance on wood panel products can be improved when a stabilising pretreatment is used. The aspen WB was a single exception in this study, and the failure observed with this panel product was primarily due to adhesive failure of the substrate. Semitransparent stains are usually not recommended for use on HB because of generally poor weathering performance. Research work at the Forest Products Laboratory and elsewhere 19,21,22,23,26,34 has shown that certain inorganic chemicals (especially hexavalent chromium compounds), when applied as dilute aqueous solutions to wood surfaces, provide the following benefits: 1. retarding degradation of wood surfaces by ultraviolet light radiation; 2. improving durability of ultraviolet light-transparent polymer coatings; 3. improving durability of paints and stains; 4. providing a degree of dimensional stabilisation to wood surfaces; 5. providing fungal resistance to wood surfaces and to coatings on the surface; 6. serving without further treatment as natural finishes for wood; 7. fixing water-soluble extractives in woods such as redwood and cedar, thereby minimising subsequent staining of applied latex paints. One of the most successful treatments investigated was that containing chromium trioxide 21 (also called chromic acid, chromic anhydride). The treatment is simple and involves dissolving 5 g of chromium trioxide ( ) in 95 g of water and applying the solution to the wood surface. Nearly all past work has been done on solid wood. The value of the pretreatment on reconstituted wood panel products is included in our recent studies. In these recent studies, the pretreatment improved the outdoor weathering performance of semitransparent pigmented finishes on a range of wood panel products (Table 8). but did not improve the performance of an opaque pigmented product (acrylic latex paint). This difference in performance resulted from the fact that most paints are not ultraviolet light-transparent whereas the semitransparent finishes are somewhat transparent to ultraviolet light 13,23,32. Thus, degradation of the wood surface occurred under the semitransparent finish. This led to loss of surface wood fibres with an accompanying loss of finish. Because the pretreatment retards degradation of the wood surface by ultraviolet light radiation, performance of semitransparent finishes is enhanced. The paint finish system performed about equally well over the pretreatment on the four panel products investigated in this study. The laboratory-prepared semitransparent finishes had their poorest performance on the aspen waferboard (WB). This is primarily caused by the delamination of surface wafers during weathering 20. The semitransparent finishes, either with or without pretreatment. must not be used for such relatively fragile surfaces.

7 WEATHERING INTERACTIONS ON TREATED AND UNTREATED WOOD SURFACES 19 TABLE 8 Role of chromium trioxide pretreatment in the performance of paints and stains over wood panel products after 6 years of outdoor exposure Finish Pretreatment Checking/ cracking Substrate Finish Overall performance rating after Flaking/ delamination Mildew Erosion General 6 Years 4 Years 2 Years Latex semitransparent stain Oil semitransparent stain 2 Coats Latex semitransparent stain Oil semitransparent stain 2 Coats Latex Pain A DOUGLAS-FIR ROUGHSAWN PLYWOOD (DFP-R) WESTERN REDCEDAR ROUGHSAWN PLYWOOD (C-P) Latex semitransparent stain Oil semitransparent stain 2 Coats Latex semitransparent stain Oil semitransparent stain 2 Coats TEXTURED HARDBOARD ASPEN WAFERBOARD (WB) The oil-based (linseed) semitransparent stain performed better than did the latex stain. The latex product actually performed like a very dilute paint, forming a thin surface film which failed like a paint. The oil finishes penetrate the wood surface to a degree. Because no discrete film forms, flaking and peeling do not occur. Finishes on Roughsawn Southern Pine Plywood This study strongly suggests that, for certain wood species, a roughsawn plywood surface may be a better surface for subsequent finish performance than that observed for the solid wood. Thus, southern pine wood would be classified as a wood difficult to finish 37, but roughsawn southern pine plywood would be classified as a good substrate to finish. Southern pine plywood is finding increased use for construction and siding (roughsawn). Because southern pine wood itself is classified as having poor characteristics for painting and finishing 37 the finishing of a plywood product made from this wood presents considerable challenge to good finish performance. In addition, plywood products are subject to face checking which leads to greater stresses being placed on filmforming finishes 13,16,20. In this study, finishes ranging from a penetrating (as a natural finish), to a film-forming clear varnish, to pigmented film-forming finishes (paints. solid colour stains) were evaluated over a four-year period on southern pine roughsawn plywood (SPP-R). The two-coat paint systems. either with or without a pretreatment, exhibited excellent performance (Table 9). Overall performance ratings dropped only to values of 9 0 to 9 4 (perfect performance would be a 10 rating). Substrate surface checking was nearly totally prevented. The acrylic latex topcoat systems (two different commercial products) performed excellently, as they did in an earlier study 16. Considering that only one coat was applied. the two solid colour stain products performed adequately. Unfortunately, substrate surface checking (face checking) progressed to the point where the finish also cracked. At this point, finish failure usually accelerates. The penetrating natural finish and the film-forming varnish gave reasonable performance for one year but declined quite rapidly thereafter. This dearly illustrates that these transparent finishes should not be used on southern pine plywood or any other plywood that will be exposed outdoors. The semitransparent oil stain is a finish that is very popular in the United States 8,29,31. This finish does give reasonably good protection and performance on SPP-R (Table 9). However, facechecking is not prevented and the finish will most likely need refinishing after about five years. This study was also conducted on southern pine scratch sanded plywood (SPP-S) but the data are not shown. Finish performance of the paints was similar on SPP-S as compared to

8 20 RECORD OF THE 1984 ANNUAL CONVENTION OF THE BRITISH WOOD PRESERVING ASSOCIATION TABLE 9 Performance of finishes on roughsawn southern pine plywood (SPP-R) after 4 years outdoor exposure Finish System Varnish 3 coats Natural finish (Linseed oil + fungicide) 1 Semitransparent oil stain Solid colour oil stain Solid colour latex stain Latex primer paint + latex topcoat Paint B + latex primer paint + latex topcoat Paint B Oil primer paint + latex topcoat Paint A + oil primer paint + latex topcoat Paint A Substrate Finish Overall performance rating after surface checking Mildew Erosion Cracking General 4 Years 3 Years 2 Years 1 Year 1 Contained 5% pentachlorophenol in a water-repellent preservative formulation SPP-R. The other finishes did not perform as well on SPP-S, primarily because the smoother surface made it more difficult to put as much finish material on the panel. The roughsawn surface is definitely a superior surface to finish with both paints and stains and usually results in better performance 16,20,22. Finishes and Pretreatments on Aspen Waferboard (WB) The study shown in Table 10 shows that pretreatments can be effective in improving finish performance on aspen waferboard only if the substrate itself does not deteriorate. Failure in the panel is not necessarily related to finish failure. The waferboard currently being produced in the United States for sheathing or underlayment is sometimes used and recommended for siding by the manufacturer. As shown in the study described above, and in earlier studies 20, this substrate does not appear to be especially durable in outdoor exposure applications. Careful finishing would be required for good outdoor performance. A question that quickly comes to mind is whether or not pretreatments would be beneficial in helping to improve the weathering performance of both this substrate TABLE 10 Role of pretreatments in the performance of paints and stains on aspen waferboard (WB) after 3 years of outdoor exposure Finish Solid colour oil stain - 1 coat Solid colour oil stain - 2 coats Solid colour latex stain - 1 coat Solid colour latex stain - 2 coats Oil primer + latex topcoat A Latex primer + latex topcoat B Substrate Finish Overall Pre- performance treatment checking Flaking Erosion Checking/ General rating cracking Wr Organo-Cr Wr Organo-Cr Wr Organo-Cr Wr Organo-Cr Wr Organo-Cr Wr Organo-Cr

9 WEATHERING INTERACTIONS ON TREATED AND UNTREATED WOOD SURFACES 21 and applied finishes, Work on particleboard products of other species has shown that pretreatments are beneficial with those materials for improving performance 33,34,35,36. The most commonly available waferboard in the United States is made from the hardwood aspen. In the study shown in Table 10, all sides and edges of the WB substrate were treated before installation and finishing on the test panel. The finishes were also applied on all edges and the front surface. The finishes and pretreatments chosen for the study shown in Table 10 were known to be useful products for solid wood. The one-coat solid colour stain finishes showed signs of fairly rapid deterioration after three years of exposure (Table 10). Application of a second coat improved performance significantly. The primary cause of poor performance was the flaking off and eventual loss of surface wafers. There were good indications that a, or chromium compound pretreatment enhanced and improved finish performance and substrate performance when the top surfaces and all edges were pretreated. Improvement in performance was observed for both one- and two-coat finishes. Finish and substrate performance were not adequate when the one-coat solid colour latex stain was used, either with or without pretreatment. Indications are that the one-coat solid colour oil stain will also prove inadequate with another year of outdoor exposure. The two-coat (primer + topcoat) paint systems performed well whether or not the WB was pretreated. Problems in either substrate or finish performance do not appear before five or six years of exposure. The study includes three-coat paint systems (one primer plus two topcoats), and there was no sign of any loss of overall performance with these systems. provide a degree of water repellency and inhibition of staining fungi. The pretreatments do not seem to have any marked effect on paint performance. Substrates The commonly used siding products, hardboard and roughsawn plywood, can he finished in ways similar to those used for solid wood. The best performance can be expected by using sufficient coats of quality pigmented finish products. Semitransparent finishes are not very good for these substrates. Less durable substrates such as aspen waferboard-may require pretreatment and careful application of quality paint systems. All these results assume that the siding products will be installed using currently recommended procedures for best performance. REFERENCES SUMMARY AND CONCLUSIONS Finishes These outdoor exposure studies of the performance of various finish systems on a variety of wood-based panel products and on some representative solid wood substrate dearly illustrate the enhanced protection found with two- and three-coat paint systems. Semitransparent pigmented stains, whether oil or latex, are partially ultraviolet light-transparent and provide less protection to the wood substrate surface than do the paint systems. Solid colour stains give a protection and performance. intermediate between paints and semitransparent stains. Unpigmented finishes (oils, water-repellent preservatives) give the least protection to the wood surfaces of all the finishes investigated. These unpigmented finishes are not suitable for use on wood panel products because of the small amount of protection they provide. Two-coat systems, comprised of acrylic latex finishes, exhibit better performance than do alkyd (oil) primer/acrylic latex paint systems. Three-coat systems (one primer plus two topcoats) always give the best overall performance and provided the greatest degree of protection for the substrate. The acrylic latex primer/acrylic latex topcoat paint systems represent the best overall finish system currently available for use on all the variety of wood substrates included in these studies. Pretreatments The water repellent and water-repellent preservative pretreatments produce variable results depending upon the substrate, the primer, and the topcoat. Except when used as a pretreatment for a pigmented shellac primer, these water repellents improve the performance of paints on solid wood. Results are variable on plywood and hardboard. The water repellent and water-repellent preservative pretreatments appear to be of some value when used on aspen waferboard. The chromium-based ptetreatments enhance panel properties and weathering performance of semitransparent pigmented stains, oil or latex. The pretreatments stabilise the wood surface against ultraviolet light degradation and also

10 22 RECORD OF THE 1984 ANNUAL CONVENTION OF THE BRITISH WOOD PRESERVING ASSOCIATION DISCUSSION ON PAPER 2 Chairman: J. David THE CHAIRMAN: Following that fascinating presentation Bill has said that he is prepared to answer some questions. Can I remind you of the ground rules. When you get up will you please state your name and your identification. This is to help the Chairman, who panics anyway, because he never recognises his best friend. Who would like to ask the first question. MR. B. JENSEN (GORI Research Limited. Denmark): I wondered if you could give some sort of explanation for the very good results you had from chromate. It has been shown in several works and studies that this type of treatment gives a very good performance. Would you give some sort of explanation of what really happens to the wood. DR. FEIST: The question is why the chromium treatment seemed to work so well on the wood surface. That is an area in which we have been interested for almost 10 years now and Tony Pizzi in South Africa has been as well. We have gone through the basic side and the practical side and it really seems to come down to the fact that chorme is fixed on the surface through a series of chemical reactions, that the presence of that chrome then acts as a U.V. blocker or controller by some mechanism, and it provides the water repellency and dimensional stabilisation of the surface which is so important for the subsequent finished performance. I think I mentioned that in our country chrome pre-treatment is not used commercially. It is very, very effective but it is not looked at because of concern environmentally. We are using the whole chromium system as a base to study, to see if we can find other materials which might perform equally. Up to now -we have found nothing that stands out like chromium. I guess that is sort of true of automobile bumpers too. Chromium plating was very popular for many decades and nothing was used as a substitute for it. The detailed mechanisms we honestly do not understand at this particular time with the exception of the water repellency, and the fact that chromium is fixed on the wood surface. Joe mentioned iron oxides. Iron oxides work as well, but they do not fix on the wood surface and if there is no protection over them they tend to be washed out very quickly. Chromium in the hexavalent state is important because it is reduced to the trivalent state and fixed. Oxides are formed, organo-metallic type complexes are formed in the wood and the treatment is there almost as a discrete layer of chromium oxide pigment. Chromium oxide pigments are very effective even on metal. In all the work we have looked at we have not found really good explanations as to the real basis of chromium s effectiveness. If we knew the mechanism I think that we could work on other chemicals which might be equally as effective. THE CHAIRMAN: Some clues if you like: chrome oxide is an extremely good infra-red reflectant and has some very interesting electroc receptive behaviour. Whether you believe oxidation occurs at the surface through peroxide for some reason or not it is an area where some quite fascinating work has been done. MR. A. C. OLIVER (Buckinghamshire College of Higher Education): I would firstly congratulate Mr. Feist on a most interesting presentation. I would ask one fairly simple question. I hope. One of the problems we have in the U.K. with painting plywood is checking of the face veneer. Is this a problem in your area in the States? DR. FEIST: Yes, the problem of face-checking is a major one in America. It is probably one of the main reasons why the use of plywood has not grown faster than it has. Work by the American Plywood Association shows that only one paint system could be used to actually control face-checking. This was an all-acrylic latex type of product, with latex primer and topcoats. It is remarkable to see a bad plywood surface facechecked within a few weeks in our climate and the effect these all-acrylic latex materials have. With any other finishing system, face-checking is quite a problem. If the plywood is not made up to standards, the face-checking can fairly quickly lead to delamination. It is the sort of thing which people expect. If you buy plywood you buy face-checking; if you buy concrete floors in our country you buy cracks. It is interesting that the newer flexible paint materials could actually prevent face-checking or at least the appearance of face-checking. It may well be that face-checking was occurring underneath the paint but you did not see it and the material retained flexibility and did not drop into that face-check as with the brittle alkyd paints or the less flexible latex materials. I think I mentioned the latex products other than the acrylics. If you were to apply a vinyl-acrylic or modified-acrylic or a polyvinyl-acetate latex material in our country, I should not perhaps say that you would be applying an inferior material, but you would be applying a material which is of poorer quality than the all-acrylic latex and you would not find their performance much beyond that of an alkyd type paint. It is only the all-acrylic systems in our country which are so outstanding. DR. C. R. COGGONS (Rentokil Limited): Returning to the issue of chromium on the surface of boards and its role in improving the performance of finishes, you mentioned the obvious problems of using simple chromium solutions for that purpose. You did mention, in passing really, that one way in which you can get chromium on to the surface is with C.C.A.

11 DISCUSSION ON PAPER 2 23 pre-treatments. I wonder if you have any data which indicates whether you get a similar performance from a normal C.C.A. pre-treatment compared with a brush applied treatment with, I think you mentioned, 5 per cent chromium trioxide solution? DR. FEIST: Yes. The whole area of chromium pretreatments resulted from our observations that in the field, painted fence posts and stained fence posts and fence posts that were just out in decay tests were performing so much better when C.C.A. treated. I do not know if anybody has specifically looked in detail at the application of finishes in a controlled system and evaluated them but we assume that it works equally as well because we took a standard formula of C.C.A. and brushed it on and let it cure either by heat or by standing and it gave nearly the same results as the final treatment with chromium trioxide as long as the chromium content was about the same. It was the chromium concentration that was most important and if you balance the chromium with the other products, the arsenic or the copper, you would not get it. In other words you have to have two and a half per cent chromium by weight in the applied finish, in the water solution. However, I have just stained my C.C.A.-treated fence posts with a simple transparent stain and I suspect they are going to look just like my test fence sample. We assume it will give the same sort of performance and a similar durability. The interesting thing is though, that these treatments will not have any real apparent effect if the finish is a totally opaque system. Opaque paints do not seem to have as marked an effect as the semi-transparent systems do and as do the totally transparent systems. John Black took a totally transparent silicone which was developed for a potting compound, I think. and smeared it on treated wood some 18 years ago or maybe 12 years ago. Those coatings are still sitting on their stabilised wood surface. A varnish will deteriorate in under three Years usually. So, finishes that will deteriorate because of U.V. deterioration are greatly enhanced by the chromium treatment. The opaque systems do not seem to have a tremendous effect. DR. J. W. W. MORGAN (Princes Risborough Laboratory): Could I follow up the question of chromium? I am sorry to dwell on this. DR. FEIST: Everybody does. Whenever I talk. if I invite comments for five minutes the questions are always about chromium. It is a very fascinating subject. DR. MORGAN: Very fascinating. You start off with chromium trioxide which presumably reduces to divalent chromium. Do you get the same effect if you apply divalent chromium directly? DR. FEIST: The hexavalent is reduced primarily to trivalent and it becomes an insoluble oxide. If you apply the trivalent chromium - this is work which has just been published-say a chromium sulphate in a plus three state - in the exposure it will tend to wash out very quickly because chromium cannot become fixed from the plus three state. However, we found that if you took a trivalent chromium nitrate and heated it. the nitrate decomposed, some chemical interaction occurred. and the chromium three was fixed and it reacted almost as if it had been chromium six initially. The key is in tying up the chromium on the wood surface in some complex form that is insoluble. We also took a product called Quillon C which is an organochrome (trivalent chromium) material made by the DuPont Company. We brushed this on the surface. heated it and some fixation occurred. It also acted as if it had &en a hexavalent chromium trioxide. So it is not necessary for the chromium to be in the hexavalent state: it is the tying up of the chromium on the wood surface so that it is insoluble to any subsequent washing and leaching that occurs that is important. We found that chromium in any form if kept on the surface will give a good performance for the subsequently applied finish. It will also give an incredible water repellency. We took a piece of filter paper, dipped it in chromium nitrate (trivalent), heated it briefly and we could not get that piece of filter paper to sink in water after several days. The water repellency was immense. It really did have a marked effect. Chromium is an incredibly unique material. THE CHAIRMAN: Quillon is dirata chromo fluoride. DR. E. R. MILLER (Princes Risborough Laboratory): I do not know whether to be heartened or depressed that your paints do not seem to be a lot better than outs. I think this must reflect just how difficult the problems are because certainly over here we think of America as being most successful in economy of labour and to have two to three year maintenance invervals even, to us, seems to be too short today. Our recent work fits in very well with yours in terms of the performance of water-borne acrylics which really do seem to be exceptionally good in terms of coating durability although they do pose problems in the joinery context. I would be interested in any views you have on more durable paints in the joinery field because we have heard now for years of certain parts of your industry which use. for instance Dupac polyurethane systems with the objective of attaining a much longer maintenance interval. I should be interested in your comments as to how successful these ate. Just before I sit down may I make a comment on the chromium issue. I think that probably the water repellency and the hydrophobic character of wood which is treated in chromates could well be one of the most important aspects because in the very limited amount of work that we have done on the irradiation of chromate treated wood under dry conditions there is very little evidence, in fact. that the presence of chromium reduces the rate of photodegradadon. DR. FEIST: Unfortunately, Roy, I cannot give you very much of an answer to the first part of your question. The joinery industry is rather tight about their information on the sort of finishes which they use. They have gone over more to the waterborne systems and they want to use the acrylic type paints but they really do not reveal much of their information at all. I suspect that urethane type paint is not used heavily and the people I have talked to at least, from the three main window manufacturing industries in our country, mostly use the acrylic type paints. However we have a very different window type system from you. We have lots of double and now triple glazing in our country and they use generally the spray and oven type treatments. I cannot give you much more of an answer than that unfortunately. DR. MILLER: How often do you paint your windows? DR. FEIST: How often do I paint my windows? I have to admit that my windows on the outside are aluminium. (Laughter). But on the inside they are all wood. My house is painted only every eight years. If you apply a goal three coat system you can get 10 years performance but only eight on the south side. By the way, that is a very common window in our country: aluminium clad wood windows with the wood being exposed indoors and the aluminium being exposed outdoors. You would have difficulty buying an all wood window any more. Vinyl cladding on wood is something which is also growing very rapidly. The manufacturers have quite a problem because they are fully encasing wood in a plastic material which is almost totally impregnable except that they have not found a good way of taking care of the exposed wood ends which is where 98 per cent of the degradation starts anyway. They are having those sorts of performance problems. It looks as if that is the sort of window which we will be having in the future. THE CHAIRMAN: Well, Ladies and Gentlemen, it is always a questionable pleasure for a Chairman to have to break off what is obviously an interesting and very enjoyable discussion period. I am afraid I have to do so as it is time for lunch. Can I say how much I, personally, have enjoyed Bill s presentation and can I ask you to join me in thanking him in the usual way. (Applause).

12 Feist, W. C. Weathering interactions on treated and untreated wood surfaces. In: Record of the 1984 Annual Convention of the British Wood Preserving Association; 1984 July 2-5; Cambridge [England]. London: British Wood Preserving Association; 1984: